to map structure–function relationships [104, 105]. Third, extraordinarily high
turnover frequencies exceeding 10
4 s
À1 have been reported [106–108]. As a
substitute for the eight-membered diphosphine, the seven-membered ligand
P
Ph
2 N
Ph
¼1,3,6-triphenyl-1-aza-3,6-diphosphacycloheptane
has
also
been
employed as a means to prevent formation of an unproductive isomer in which a
proton is pinched between the two amines of a single ligand (Fig. 14b,c). The
compound [Ni(P
Ph
2 N
Ph ) 2 ] is reported to have a turnover frequency of more than
100,000 s
À1 under optimal conditions [106]. However, the change from an eightmembered ligand to a seven-membered one also has an impact on coordination
geometry, enforcing a more planar geometry about the nickel (Fig. 14c). The
change in geometry is likely to be the cause of the increase in required overpotential
that accompanies the increase in rate for the P
Ph
2 N
Ph complex. This is a reminder
that changing one aspect of a complex can, and often does, have unintended
consequences on other properties, and predictive design of molecular catalysts is
by no means a mature field.
Due to their exceptional turnover frequencies as homogeneous electrocatalysts,
several groups have reported efforts to immobilize mononuclear nickel catalysts on
electrodes to create heterogeneous systems. The Bullock group recently reported an
ester-functionalized variant of P
R
2 N
R
0
2 which they could use as a handle to covalently functionalize glassy carbon electrodes. Although the functionalized electrodes have catalytic properties similar to related compounds in solution, electrode
immobilization results in markedly increased instability under acidic conditions
[109]. Promisingly, Artero and coworkers reported the non-covalent immobilization of a pyrene-functionalized version of a member of this family on carbon
nanotubes resulting in electrodes with current densities of 20 mA cm
À2 for hydrogen evolution [110]. In related work, covalent attachment to the carbon nanotubes
resulted in exceptionally stable catalysts, and turnover number reported to be more
than 100,000, requiring only 10s of mV of overpotential, but the current density was
relatively low [111]. Thus, it remains a challenge to translate the knowledge from
homogeneous catalysis to create a related heterogeneous system that is fast, efficient, and stable.
Shaw and coworkers have explored the ability to tune proton reduction catalysis
by modifications of the second coordination sphere starting from the nickel phosphine complexes of DuBois described above. They have demonstrated that modification of the outer coordination sphere of [Ni((P 2 N 2 R)-N-R
0 ) 2 ]
2+ complexes with
amino acids can significantly modify catalytic properties (Fig. 14d)
[112, 113]. Remarkably, several derivatives are capable of fully reversible H 2
production and oxidation in aqueous solutions with pH values in the range of 0–6
[114]. By comparing the rates of complexes with various amino acids, they have
hypothesized that a carboxylic acid in the outer coordination sphere may enhance
catalysis, much like proton transfer residues in natural enzymes [115].
First coordination spheres including nitrogen and sulfur donors have also been
used to construct highly active nickel proton reduction catalysts. Employing a
polypyridyl ligand framework, Sun and coworkers have created the complex [Ni
(L) 2 (H 2 O) 2 ](BF 4 ) 2 for L¼2-(2-pyridyl)-1,8-naphthyridine, which, under optimal,
basic, photocatalytic conditions, has a remarkable turnover number of 3230
Biomimetic Complexes for Production of Dihydrogen and Reduction of CO 2
251
Précédent

- 261/287

Suivant